• Journal of information and communication convergence engineering
• /
• v.20 no.2
• /
• pp.103-112
• /
• 2022

The goal of pattern mining is to identify novel patterns in a database. High utility itemset mining (HUIM) is a research direction for pattern mining. This is different from frequent itemset mining (FIM), which additionally considers the quantity and profit of the commodity. Several algorithms have been used to mine high utility itemsets (HUIs). The original BPSO algorithm lacks local search capabilities in the subsequent stage, resulting in insufficient HUIs to be mined. Compared to the transfer function used in the original PSO algorithm, the V-shaped transfer function more sufficiently reflects the probability between the velocity and position change of the particles. Considering the influence of the acceleration factor on the particle motion mode and trajectory, a nonlinear acceleration strategy was used to enhance the search ability of the particles. Experiments show that the number of mined HUIs is 73% higher than that of the original BPSO algorithm, which indicates better performance of the proposed algorithm.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.9
• /
• pp.371-380
• /
• 2022

Convolutional neural networks are widely used to manipulate data arranged in a grid, such as images. A general convolutional neural network consists of a convolutional layers and a fully connected layers, and each layer contains a nonlinear activation functions. This paper proposes a combined parametric activation function to improve the performance of convolutional neural networks. The combined parametric activation function is created by adding the parametric activation functions to which parameters that convert the scale and location of the activation function are applied. Various nonlinear intervals can be created according to parameters that convert multiple scales and locations, and parameters can be learned in the direction of minimizing the loss function calculated by the given input data. As a result of testing the performance of the convolutional neural network using the combined parametric activation function on the MNIST, Fashion MNIST, CIFAR10 and CIFAR100 classification problems, it was confirmed that it had better performance than other activation functions.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.6 s.177
• /
• pp.1593-1600
• /
• 2000

A method based on frequency domain approaches is presented for the nonlinear parameters identification of structure having nonlinear joints. The finite element model of linear substructure is us ed to calculating its frequency response functions needed in parameter identification process. This method is easily applicable to a complex real structure having nonlinear elements since it uses the frequency response function of finite element model. Since this method is performed in frequency domain, the number of equations required to identify the unknown parameters can be easily increased as many as it needed, just by not only varying excitation amplitude but also selecting excitation frequencies. The validity of this method is tested numerically and experimentally with a cantilever beam having the nonlinear element. It was verified through examples that the method is useful to identify the nonlinear parameters of a structure having arbitary nonlinear boundaries.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.17 no.8
• /
• pp.747-756
• /
• 2007

The hydropneumatic suspension units, which have applied to the tracked vehicles, have the spring and damping function in the unit. The nonlinear characteristics such as roadwheel rotation effects, gas behavior changes, hydraulic damping characteristics, hysterisis, and frictional forces have been ignored or simplified to analyze the mathematical models in many areas. This study describes the dynamic characteristics and the nonlinear behaviors of hydropneumatic suspension unit considering the nonlinear effects such as the nonlinear spring and nonlinear damping through the simulation and the experiment. The utility of nonlinear analysis through the higher-order spectral analysis is also presented.

• Korean Journal of Mathematics
• /
• v.17 no.4
• /
• pp.507-519
• /
• 2009

We give a theorem of the existence of the multiple solutions of the Hamiltonian system with the square growth nonlinearity. We show the existence of m solutions of the Hamiltonian system when the square growth nonlinearity satisfies some given conditions. We use critical point theory induced from the invariant function and invariant linear subspace.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.95.4-95
• /
• 2001

In this paper, we proposed and analyze an robust adaptive control scheme for uncertain nonlinear systems using Universal function approximators. The proposed scheme completely overcomes the singularity problem which occurs in the indirect adaptive feedback linearizing control. No projection in the estimated parameters and no switching in the control input are needed. The stability of the closed-loop systems is guaranteed in the Lyapunov standpoint.

• Journal of applied mathematics & informatics
• /
• v.36 no.3_4
• /
• pp.271-283
• /
• 2018

In this paper, we introduce various first order (p, q)-difference equations. We investigate solutions to equations which are linear (p, q)-difference equations and nonlinear (p, q)-difference equations. We also find some properties of (p, q)-calculus, exponential functions, and inverse function.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.214-217
• /
• 1999

In this paper, genetic algorithm (GA) is implemented to search for the optimal structures (i.e. the kind of neural networks, the number of inputs and hidden neurons) of neural networks which are used approximating a given nonlinear function. Two kinds of neural networks, i.e. the multilayer feedforward [1] and time delay neural networks (TDNN) [2] are involved in this paper. The synapse weights of each neural network in each generation are obtained by associated training algorithms. The simulation results of nonlinear function approximation are given out and some improvements in the future are outlined.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2000.10a
• /
• pp.135-139
• /
• 2000

This research proposes a nonlinear sliding mode control. The sliding mode control is designed according to Lyapunov function. The equivalent control term is estimated by neural network. To estimate the unknown part in the control law in on-line fashion, A recurrent neural network is given as on-line estimator. The stability of the control system is guaranteed owing to the on-line learning ability of the recurrent neural network. It is certificated through simulation results to be applied to nonlinear system that the function approximation and the proposed control scheme is very effective.

• Proceedings of the KIEE Conference
• /
• 1998.07c
• /
• pp.973-975
• /
• 1998

For a nonlinear feedback linearization-full order observer/sliding mode controller (NFL-FOO/SMC), the separation principle is derived, and the closed-loop stability is proved by a Lyapunov function candidate using an addition form of the sliding surface vector and the estimation error.